Raw LLM Responses

Inspect the exact model output for any coded comment.

Comment
If the people instructing the ai don't know how to properly develop software (aka 'junior developers') of course you get crap. I've seen what would take half a day to do manually done in minutes when ai is instructed by someone who knows how systems are built, with the ai producing well structured, complete and reliable code - based on their instructions. The idea of 'vibe coding' is the antithesis of how to use these tools to produce good systems. The quality of what you get is based on the quality and completeness of what you ask for. No different than someone giving incomplete instructions to a developer, you have no idea what you will get back. That includes knowing how to structure the system itself and its underlying architecture. Unless of course your building a tic tac toe app. Problem is most developers don't know how to write complete and proper requirements and specifications. You don't need to write code yourself anymore, but you do know how to review code and build systems as well as writer proper requirements and specifications. Nothing has changed in this regard, you get what you ask for has always been the case when asking a developer to write code. Generalizing on low level junior developers vibe coding is nonsense. A senior engineer who understand the systems requirements and able to write solid requirements can today do the work of a small dev team. Anyone who hasn't gotten significant returns in using ai to develop software, doesn't know how to build robust and resilient software. If you are thinking about 'coding', you need to first learn what 'software engineering' means before speaking about ai as a tool in the software development lifecycle. Coming from decades of experience in the space and worked with the full spectrum of developers from genius to clueless/useless. If you don't know how to design and build systems, how can you expect to instruct a person or ai to build something proper, reliable, resilient, scalable and maintainable. Nothing has changed in this regard, its only been brought out into the open.
youtube AI Jobs 2026-02-04T11:2… ♥ 1
Coding Result
DimensionValue
Responsibilityuser
Reasoningconsequentialist
Policynone
Emotionapproval
Coded at2026-04-26T23:09:12.988011
Raw LLM Response
[ {"id":"ytc_UgzWp-s58ZrKfdYDmwR4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytc_UgxT2-v7D2tcuWxzpPV4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"}, {"id":"ytc_UgxHkt1cV-NdV6QZxt14AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"indifference"}, {"id":"ytc_UgxDc81VvJuZBs88efZ4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytc_UgynUg7xE-ftWl0kObZ4AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"}, {"id":"ytc_UgxrnEUnEY-Dz9m2wM14AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"}, {"id":"ytc_UgzNAe88YxjQZbQhWF14AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytc_Ugx5Ysxv3uArT2iInPl4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytc_Ugz0-uM8832rGEVFS9d4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"approval"}, {"id":"ytc_Ugx1MA3obtCOVtGzZ8N4AaABAg","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"outrage"} ]